复杂运动目标逆合成孔径雷达成像关键技术研究

Inverse synthetic aperture radar can provide the valuable target information of shape, size, structure and attitude. At present, ISAR imaging has been an important technique for target recognition. However, the targets are usually non-cooperative, which means that the targets may move in different patterns of motion, such as hypersonic moving, rapidly spinning, precession, three dimensional rotating, etc. The traditional imaging method is not effective in non-cooperative target imaging: this has greatly inhibited the application of ISAR technique to the automatic target recognition, and makes the imaging study for targets with complex motion become an urgent problem. In order to obtain a focused ISAR image, several special difficult problems need to be resolved when observing targets with complex motion. The problems includes the motion compensation in high resolution range profile (HRRP) imaging, Doppler focalization for rapidly spinning targets, Doppler focalization for precession targets, Doppler focalization for three dimensional rotating targets, and Doppler focalization for targets with rotating parts. In this research, based on the analyses results of the simulated and the anechoic chamber data, the higher accuracy signal models which can describe the electromagnetic characters of targets with complex motion are built, respectively. Then, the resolutions for overcoming above-mentioned problems will be proposed based on the built models. In the research process, some real data will be used for testing the validity of the proposed algorithms. According to the test results, the proposed algorithms will be improved time after time. The research results achieved in this program will help for improving the performance of the ISAR imaging for targets with complex motion, and promotes the practical application process of the proposed ISAR technique.

逆合成孔径雷达成像能够获取目标大小、形状、结构及姿态等细节信息，是目标识别的一种重要技术手段。然而，由于观测目标是非合作的，通常具有高速运动、机动、自旋和进动以及三维转动等复杂运动形式，严重制约了逆合成孔径雷达成像在目标识别领域的应用，加强复杂运动目标逆合成孔径雷达成像的研究成为迫切需求。复杂运动目标逆合成孔径雷达成像需要解决一维距离像运动补偿、自旋/进动目标、三维旋转目标、以及带旋转部件目标的多普勒聚焦等特殊难点问题。为此，本项目从仿真和暗室测量数据分析着手，建立能够准确描述复杂运动目标特性的宽带雷达回波模型，然后基于所建模型提出针对上述难点问题的解决方法。研究过程中，将利用复杂运动目标雷达实测数据对所研究算法的有效性进行反复验证，并根据验证结果对所提算法进行改进和完善。本项目研究将为提高逆合成孔径雷达对复杂运动目标的成像精度奠定基础，推动该项技术的实用化进程。

逆合成孔径雷达成像能够获取目标大小、形状、结构及姿态等细节信息，是目标识别的一种重要技术手段。然而，由于观测目标是非合作的，通常具有高速运动、机动、自旋和进动以及三维转动等复杂运动形式，严重制约了逆合成孔径雷达成像在目标识别领域的应用，加强复杂运动目标逆合成孔径雷达成像的研究成为迫切需求。.本项目针对几类典型复杂运动分别开展了ISAR成像算法的研究。主要工作总结如下：首先，针对目标高速运动导致传统调频步进信号距离像散焦的情况，提出一种脉冲重复时间可变的调频步进信号，通过对脉冲重复时间预先设计来消除目标速度与频率的耦合效应，并通过回波对消技术消除目标加速度对合成距离像的影响，给出了该算法对脉冲重复时间的精度要求，从而高效地实现了运动目标的高分辨成像问题；其次，针对自旋/进动目标，在准确建立其回波模型的基础上，对目标的自旋/进动参数进行估计，并在此基础上实现目标的多普勒维聚焦。再次，针对三维转动目标的成像方法进行了研究，建立起回波模型，提出基于能量累积的成像时间段选取方法，该算法具有运算量较小的优点。通过选择最佳成像时间，保证在较长的相干积累时间内复杂运动近似为匀速转动，从而利用RD算法进行有效成像接着对包含旋转部件的目标，采用小波变换方法实现对主题回波与旋转部件回波的有效分离，然后分别对两个部分进行常规ISAR成像。最后，根据观测条件的不同，提出了两种目标三维超分辨成像方法。研究过程中，利用仿真和暗室测量数据对所研究算法的有效性进行反复验证，并根据验证结果对所提算法进行改进和完善。本项目研究将为提高逆合成孔径雷达对复杂运动目标的成像精度奠定基础，推动该项技术的实用化进程。